DoC2k_init problem

DoC2k_init problem

[Patrick Higgins]

The problem seems to be that on line 392 of doc2000,

old = (struct DiskOnChip *)old->nextdoc->priv;

old->nextdoc is NULL.  What I don't seem to understand is why this old
list is being searched at all.  It should only be searched if doc2klist is
not NULL, and I see no reason why it should be non-NULL on system bootup,
unless it's static initializer isn't actually working, or a buffer is
being overrun somewhere.

Anyone know what might be going on here?

Patrick



To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: DoC2k_init problem

[David Woodhouse]

phiggins@transzap.com said:
>  Anyone know what might be going on here?

Pure stupidity on my part. Thanks for fixing it .

The reason we're searching through the old list is to make sure that we 
don't get invoked more than once for a DiskOnChip in any given location - 
which could happen if you have docprobe.o as a module, and load it more 
than once. 

There's no equivalent to request_region() for memory space, unfortunately.

--
dwmw2




To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: DoC2k_init problem

[Alessandro Rubini]

> There's no equivalent to request_region() for memory space, unfortunately.

Hmm.. what about request_mem_region()?

/alessandro


To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: DoC2k_init problem

[David Woodhouse]

rubini@pop.systemy.it said:
>  Hmm.. what about request_mem_region()?

That's only in 2.3, though, isn't it? I'm trying to avoid adding too much
2.3-specific code at this stage.

Besides which, we've just realised (and I'm writing that response now) that 
we have to eliminate duplicates more strictly than that - if you don't 
decode all the address lines, the same DiskOnChip can turn up at different 
locations too.

--
dwmw2




To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org